How can I retrieve loguru descriptive traceback?

[adhadseKavida]

I want to retrieve the loguru descriptive traceback when doing the serialized logging.

Is there is a function like traceback.format_exception(*error) but from loguru how do I get the descriptive traceback.

Here is my current setup:

import json
import sys
import traceback

import loguru
from loguru import logger

PIPELINE_NAME = "test" 
SERIALIZED_LOGGING = True

def serialize(record):
    """Serialize to our custom needs for structured logging.

    With `logger.exception(msg)` all fields of `exception` key is populated.
    With `logger.error(msg)` `traceback` field remains unpopulated.
    Anything passed as kwargs will also be appended to the log record.

    Additionally, if logger.x("@message", show_exception_value=False) will not
    serialize exception's original value and will make it None. Useful in case of
    BulkWriteError which throws whole duplicate document.

    Args:
        record: a dict record of loguru
    Returns:
        a JSON encoded str.
    """
    error: loguru.RecordException = record["exception"]
    error_by_default = sys.exc_info()  # logger.error
    pipeline: str | None = record["extra"].get("pipeline", None)
    show_exception_value: bool = record["extra"].get("show_exception_value", True)
    level: str = record["level"].name
    extra = record["extra"]
    extra.update({"pipeline": pipeline})
    if error:  # only set when exception.
        exception = {
            "type": error.type.__name__,
            "value": str(error.value) if show_exception_value else None,
            "traceback": "".join(traceback.format_exception(*error)),
        }
    elif error_by_default[0]:  # whenever error occurs
        _type, _value, _ = sys.exc_info()
        exception = {
            "type": _type.__name__,
            "value": str(_value) if show_exception_value else None,
            "traceback": None,
        }
    else:
        exception = None

    to_serialize = {
        "level": level,
        # "time": record["time"].strftime("%d-%m-%YT%H:%M:%SZ"),
        "message": record["message"],
        "pipeline": pipeline,
        "exception": exception,
    }
    to_serialize.update(extra)
    return json.dumps(to_serialize)

def patching(record):
    """Patch the logger."""
    record["extra"]["serialized"] = serialize(record)

def get_contextualized_logger(
    pipeline_name: str = PIPELINE_NAME, default_logger=logger
):
    """Generates a contextualized logger with pipeline_name (P1/P2/P3)."""
    if not SERIALIZED_LOGGING:
        return default_logger

    default_logger.remove()
    default_logger = default_logger.patch(patching)
    default_logger.add(
        sink=sys.stderr,
        serialize=False,  # custom serialization requires this to be False
        backtrace=True,
        diagnose=False,  # enable for debugging; shows runtime vals of call func
        level="INFO",
        format="{extra[serialized]}",
    )
    return default_logger

[Delgan]

There is currently no way to retrieve the formatted traceback. :/

Maybe you can use better_exceptions instead form which Loguru formatting is derived?

[adhadseKavida]

There is currently no way to retrieve the formatted traceback. :/

Maybe you can use better_exceptions instead form which Loguru formatting is derived?

I'll check it out soon and update it. Thanks!